这是我关于Stake溢出的第一篇文章，如果需要更多信息，请问我以下问题！
情况：我整理了Maritimes(加拿大大西洋(淡水生态系统的水化学数据，因为我正试图使用入侵物种的随机森林模型(RFM(创建一个预测物种分布的模型。不幸的是，加拿大大西洋缺乏一致的水监测计划，现有的监测计划也没有像其他小组一样监测相同的参数。因此，我的数据库(包括培训和测试(有许多NA。
问题：这是我从RFM:得到的回复
> p1 <- predict(model2, newdata=Test_Dataset,type="prob")[,2]
> p1
1 2 3 4 5 6 7 8 9 10 11 13 14 15 16 17 19 20 21 23 24 25 26 28 29 30 31 32 33 34不，不，不35 36 37NA NA NA
我尝试过的：
我使用各种预测因子构建了RFM(即模型2(。我确实包括：
型号2<-randomForest(CMS ~ Lat+Lon+pH+碱度+Ca+硬度+DO+TOC+T_P+T_N+Cond+Na+No_Stocking+No_Fish_Species+Dist_Hwy+No_Boat_Launches+Connected_Lakes+Invasives，重要性=TRUE，数据=TrainSet，Na.action=Na.roughfix(型号2
**注意，变量的大列表是预测因素，CMS是物种。
我尝试将测试数据集(test_dataset(与训练数据集(Validation_dataset(进行匹配。
测试数据集<-rbind(验证数据集[1，]，验证数据集(测试数据集<-测试数据集[-1，]
我搜索并阅读了多个资源(包括明显的R页面和链接到那里的参考资料(。
我对数据帧进行了如下突变(我只显示Validation_Dataset，因为两者的突变相同(：
突变数据集以修复R读取NA细胞的问题
验证数据集<-验证数据集%>%dplyr:：突变(#将年份转换为分类变量年=因子(年(，#将叶绿素浓度从字符文件转换为数字文件#在适当的时候将"NA"转换为缺失值数据叶绿素=dplyr:：na_if(叶绿素，"na"(，叶绿素=因子(叶绿素(，硬度=dplyr:：na_if(硬度，"na"(，硬度=系数(硬度(，碱度=dplyr:：na_if(碱度，"na"(，碱度=因子(碱度(，Ca=dplyr:：na_if(Ca，"na"(，Ca=因子(Ca(，TOC=dplyr:：na_if(TOC，"na"(，TOC=因子(TOC(，Cond=dplyr:：na_if(Cond，"na"(，Cond=因子(Cond(，Na=dplyr:：Na_if(Na，"Na"(，Na=因子(Cond(，NH4=dplyr:：na_if(NH4，"na"(，NH4=因子(NH4(，NO3=dplyr:：na_if(NO3，"na"(，NO3=因子(NO3(，pH=dplyr:：na_if(pH，"na"(，pH=因子(pH(，T_N=dplyr:：na_if(T_N，"na"(，T_ N=因子(T_N(，T_P=dplyr:：na_if(T_P，"na"(，T_ P＝因子(T_P(，DO=dplyr:：na_if(DO，"na"(，DO=因子(DO(，盐度=dplyr:：na_if(盐度，"na"(，盐度=因子(盐度(，No_Stocking=dplyr:：na_if(No_Stocking，"na"(，No_ Stocking=因子(No_Stocking(，No_Fish_Species=dplyr:：na_if(No_Fish_Stpecies，"na"(，No_Fish_Species=因子(No_Fish_Stpecies(，Dist_Hwy=dplyr:：na_if(Dist_Hwy，"na"(，Dist_Hwy＝因子(Dist_Hwy(，No_Boat_Launche=dplyr:：na_if(No_Boat_Launches，"na"(，发射次数=因子(发射次数(，Connected_Lakes=dplyr:：na_if(Connected_Lates，"na"(，Connected_Lakes=因子(Connected_Lates(，Invasives=dplyr:：na_if(Invasives，"na"(，Invasives=因子(Invasives(，Lat=因子(Lat(，Lon=因子(Lon(，CMS=因子(CMS(
问题：有人知道如何真正使编码工作，以便model2在Test_Dataset上进行预测吗？我认为这个问题实际上可能很小，但我没有看到
以下是训练数据集(Validation_dataset(的一瞥：
> str(Validation_Dataset)
Classes ‘spec_tbl_df’, ‘tbl_df’, ‘tbl’ and 'data.frame': 37 obs. of 31 variables:
$ Name : chr "Canard River" "Cedar Creek" "Holland River" "Speed River" ...
$ STN #/COUNTY : chr "10000200202" "16001800202" "3007700202" "16018403402" ...
$ Province : chr "ON" "ON" "ON" "ON" ...
$ Lat : Factor w/ 37 levels "42.03204214",..: 2 1 11 9 10 8 7 5 6 3 ...
$ Lon : Factor w/ 37 levels "-83.01879548",..: 1 2 11 8 10 6 7 9 5 4 ...
$ Year : Factor w/ 9 levels "2007, 2011","2010, 2015, 2011",..: 8 8 8 8 8 8 8 8 8 8 ...
$ Month : chr "4" "4" "4" "4" ...
$ Day : chr "11" "12" "26" "27" ...
$ Data Source : chr "ON Provincial (Streams) Water Quality Monitoring Network" "ON Provincial (Streams) Water Quality Monitoring Network" "ON Provincial (Streams) Water Quality Monitoring Network" "ON Provincial (Streams) Water Quality Monitoring Network" ...
$ pH : Factor w/ 35 levels "6.073333","6.13",..: 18 21 28 29 25 34 30 32 19 26 ...
$ Alkalinity : Factor w/ 31 levels "1.8","2.8","3.933333333",..: 19 22 31 30 27 NA NA 26 NA 21 ...
$ Hardness : Factor w/ 13 levels "14.8","36.8",..: 7 8 11 10 9 NA NA 13 NA NA ...
$ Ca : Factor w/ 24 levels "3.833333333",..: 18 19 24 20 21 NA NA 22 NA NA ...
$ Chlorophyll : Factor w/ 15 levels "0.423601","0.453791",..: NA NA NA NA NA NA NA NA NA NA ...
$ DO : Factor w/ 26 levels "0.27","6.2","6.96",..: 21 24 18 16 4 25 17 14 2 7 ...
$ TOC : Factor w/ 3 levels "4.8","5.5","8.8": NA NA NA NA NA NA NA NA NA NA ...
$ T_P : Factor w/ 24 levels "0.002","0.003",..: 23 22 18 10 15 14 16 13 21 20 ...
$ T_N : Factor w/ 32 levels "0.006","0.13",..: 30 31 27 28 17 29 24 32 21 25 ...
$ NO3+NO2 : num 2.173 2.292 1.092 1.695 0.426 ...
$ NO3 : Factor w/ 32 levels "0.027","0.035",..: 30 31 26 27 11 29 24 32 22 8 ...
$ NH4 : Factor w/ 27 levels "0.005","0.006",..: 26 25 22 17 9 11 13 19 23 27 ...
$ Cond : Factor w/ 34 levels "41","97","134",..: 24 21 29 23 22 14 34 31 21 17 ...
$ Salinity : Factor w/ 9 levels "0.11","0.15",..: NA NA NA NA NA NA NA NA NA NA ...
$ Na : Factor w/ 34 levels "41","97","134",..: 24 21 29 23 22 14 34 31 21 17 ...
$ No_Stocking : Factor w/ 3 levels "0","1","2": 1 2 2 3 1 2 1 2 1 2 ...
$ No_Fish_Species : Factor w/ 9 levels "0","1","2","3",..: 1 4 6 4 1 5 1 9 1 9 ...
$ Dist_Hwy : Factor w/ 16 levels "0.003","0.006",..: NA NA 16 NA NA NA NA 8 NA 5 ...
$ No_Boat_Launches: Factor w/ 8 levels "0","1","2","3",..: 1 1 5 1 1 1 1 8 1 3 ...
$ Connected_Lakes : Factor w/ 11 levels "0","1","2","3",..: 7 2 3 4 9 6 2 3 2 5 ...
$ Invasives : Factor w/ 3 levels "0","1","2": NA NA NA NA NA NA NA NA NA NA ...
$ CMS : Factor w/ 2 levels "NO","YES": 2 2 2 2 2 2 2 2 2 2 ...
使用参数na.roughfix。如果要使用它，必须首先在randomForest函数外部指定。我将以虹膜数据集为例。
iris.roughfix <- na.roughfix(iris.na)
iris.narf <- randomForest(Species ~ ., iris.na, na.action=na.roughfix)